By Deborah Borfitz
November 12, 2019 | Advances in microfluidic cell sorting techniques at the University of Washington (UW) is allowing for highly precise isolation and analysis of tumor-derived rare cells and extracellular vesicles (EVs) at high- throughput speed, according to chemistry and bioengineering professor Daniel T. Chiu, a speaker at the 2019 Next Generation Dx Summit in Washington, D.C.
It can be “tricky” to pull hundreds of rare cells out of the billions in the background, says Chiu, but the two-stage sorting process (DOI: 10.1021/acs.analchem.9b03690) developed by UW researchers is significantly improving the odds of success. It builds on their next-generation ensemble-decision aliquot ranking (eDAR) method for the sorting step that gets the job done faster and with improved reproducibility and stability.
The newly described sequential eDAR platform involves two rounds of cell sorting with fluid elements being “stretched” in between using the system’s herringbone design features, Chiu explains. The platform can be used to collect single circulating tumor cells in a 96-well plate for downstream analysis.
The objects of interest are very small—on the order of 50 to 100 nanometers—so conventional methods won’t do, Chiu says. Microfluidics employ filters to get rid of red blood cells so retained white blood cells can be color-labeled and biomarkers of interest bleached. Two-stage flow sorting recovers the scattering information that otherwise gets lost in the traditional cell sorting process, he adds.
Finding good biomarkers is always challenging, Chiu continues, and how many of them are needed to isolate cells varies across instruments. In a side-by-side analysis, the latest eDAR platform compared favorably with other cell sorting approaches, pulling out one cell per 8 milliliters of blood.
WU researchers are also interested in single-molecule detection within individual EVs, says Chiu. One of the problems in doing that with microRNA is that a billion EVs would be required to have enough mass. A second challenge is in detecting EVs when they’re not the brightest object in the mix and a third is that they come in a variety of sizes.
Ultra-fast sorting solves these problems in a couple nanoseconds, says Chiu. “We can sort between 10,000 EVs per second, and maybe hundreds of thousands per second if everything is perfect.”
The bottleneck is in flow cytometry, he says. “The cell has a lot of markers on it.” EVs with links to various types of cancers are also relatively few, ranging from a couple to perhaps 20.
EVs can be isolated using light scattering analysis, but at a much lower flow throughput, says Chiu. Another option is to “play a trick with optics,” using the intensity of reflected light off stained EVs to get at size. “That kind of works but you have to be careful because dyes [can be unpredictable]. The system has to be really stable, and you may need to do overnight things.”
The system under construction at WU involves conjugating antibody proteins into nanoparticles, says Chiu, noting that some proteins are “quite variable” EV to EV. Ultra-bright fluorescent probes permit fast interrogation, as does using concentrated serum samples, he adds.
With the next upgrade, up to 16 colors will be detectable as researchers work toward their goal of subtyping all EVs. “We’re also looking at the micro-DNA content of EVs, but that’s another story.”